Moderately elevated blood pressure during pregnancy and odds of hypertension later in life: the POUCHmoms longitudinal study.

OBJECTIVE: Hypertensive disorders in pregnancy signal an increased risk of cardiovascular disease for women. However, future hypertension risk among pregnant women with moderately elevated blood pressure (BP) is unknown. We examined associations among moderately elevated BP or hypertensive disorders during pregnancy and later prehypertension or hypertension. DESIGN: Longitudinal cohort study. SETTING: Five communities in Michigan, USA. SAMPLE: Data are from pregnant women enrolled in the Pregnancy Outcomes and Community Health Study. We included 667 women with gestational BP measurements who participated in the POUCHmoms Study follow-up 7-15 years later. METHODS: Moderately elevated BP was defined as two measures of systolic BP ≥ 120 mmHg or diastolic BP ≥80 mmHg among women without a hypertensive disorder. Weighted multinomial logistic regression models estimated odds of prehypertension or hypertension at follow-up, adjusted for maternal confounders and time to follow-up. MAIN OUTCOME MEASURES: Prehypertension or hypertension. RESULTS: Women meeting the moderately elevated BP criteria (64%) had significantly higher odds of hypertension at follow-up (adjusted odds ratio 2.6; 95% confidence interval 1.2-5.5). These increased odds were observed for moderately elevated BP first identified before or after 20 weeks of gestation, and for elevated systolic BP alone or combined with elevated diastolic BP. CONCLUSIONS: Moderately elevated BP in pregnancy may be a risk factor for future hypertension. Pregnancy offers an opportunity to identify women at risk for hypertension who may not have been identified otherwise. TWEETABLE ABSTRACT: Moderately elevated blood pressure in pregnancy may be associated with hypertension later in life.

Survival of HIV-positive individuals with hepatitis B and C infection in Michigan.

We sought to estimate mortality and associated factors in HIV-hepatitis co-infected individuals in Michigan using a retrospective cohort study. For the study period of 1 January 2006 to 31 December 2009, all HIV-infected individuals were matched to hepatitis B and C cases. In the final Cox proportional hazards regression model, individuals of other [hazard ratio (HR) 2·2, 95% confidence interval (CI) 1·4-3·2] and black (HR 1·3, 95% CI 1·1-1·6) race had decreased survival compared to white race. Similarly, injecting drug users (IDUs) (HR 2·1, 95% CI 1·6-2·6), men who have sex with men (MSM)/IDUs (HR 1·5, 95% CI 1·1-2·2), individuals with undetermined risk (HR 1·5, 95% CI 1·2-1·9) and heterosexual practices (HR 1·4, 95% CI 1·1-1·8) had decreased survival compared to MSM. Additionally, an interaction was found between current HIV status and co-infection. Mortality in HIV-hepatitis co-infected individuals remains a continuing problem. Our study can help in planning interventions to reduce mortality in HIV-infected individuals.

Hepatitis B and C co-infection in HIV/AIDS population in the state of Michigan.

A retrospective cohort study was conducted from 1 January 2006 to 31 December 2009 in Michigan to estimate the prevalence of HIV and hepatitis co-infection and identify associated factors. The prevalence of co-infection was 4.1% [95% confidence interval (CI) 3.8-4.5]. Multivariable logistic regression analysis revealed a significant association between co-infection and being male and: of Black race [odds ratio (OR) 2.0, 95% CI 1.2-3.6] and of Other race (OR 3.5, 95% CI 1.7-7.0) compared to Hispanic race. A significant association was found between co-infection and risk categories of blood products (OR 11.1, 95% CI 6.2-20.2), injecting drug user (IDU) (OR 3.6, 95% CI 2.7-4.8) and men who have sex with men/IDU (OR 3.4, 95% CI 2.4-4.9) in addition to two interactions; one between sex and current HIV status and the other between current HIV status and age at HIV diagnosis. Our results document the changing epidemiology of HIV-hepatitis co-infection which can guide preventive measures and interventions to reduce the prevalence of hepatitis co-infection.

Bayesian latent variable models for spatially correlated tooth-level binary data in caries research.

Analysis of dental caries is traditionally based on aggregated scores, which are summaries of caries experience for each individual. A well-known example of such scores is the decayed, missing and filled teeth or tooth surfaces index introduced in the 1930s. Although these scores have improved our understanding of the pattern of dental caries, there are still some fundamental questions that remain unanswered. As an example, it is well believed among dentists that there are spatial symmetries in the mouth with respect to caries, but this has never been evaluated in a statistical sense. An answer to this question requires the analysis to be performed at subunits within the mouth, which necessitates the use of methods for correlated data. We propose a Bayesian generalized latent variable model coupled with an undirected graphical model to investigate the unique spatial distribution of tooth-level caries outcomes in the mouth. Data from the Signal Tandmobiel(®) study in Flanders, a dental longitudinal survey, are used to illustrate the methodology.

A global sensitivity test for evaluating statistical hypotheses with nonidentifiable models.

We consider the problem of evaluating a statistical hypothesis when some model characteristics are nonidentifiable from observed data. Such a scenario is common in meta-analysis for assessing publication bias and in longitudinal studies for evaluating a covariate effect when dropouts are likely to be nonignorable. One possible approach to this problem is to fix a minimal set of sensitivity parameters conditional upon which hypothesized parameters are identifiable. Here, we extend this idea and show how to evaluate the hypothesis of interest using an infimum statistic over the whole support of the sensitivity parameter. We characterize the limiting distribution of the statistic as a process in the sensitivity parameter, which involves a careful theoretical analysis of its behavior under model misspecification. In practice, we suggest a nonparametric bootstrap procedure to implement this infimum test as well as to construct confidence bands for simultaneous pointwise tests across all values of the sensitivity parameter, adjusting for multiple testing. The methodology's practical utility is illustrated in an analysis of a longitudinal psychiatric study.

Moving from theory to practice: implementing the Kin Keeper Cancer Prevention Model.

This paper presents the rationale and findings of a feasibility and process study of the Kin Keeper(SM) Cancer Prevention Intervention. An observational cohort study design was implemented with African-American women in synergistic female family relationships. Community health workers (CHWs) from two Michigan public health programs recruited women to serve as 'kin keepers' who in turn recruited their female family members. In total, 161 kin keepers and female family members were sampled. Trained CHWs led kin keepers and family members in learning about breast cancer. Data methods included baseline and post-training administration of a breast cancer literacy assessment, post-training focus groups and review of personal action plans. To validate the feasibility of the process, a linear mixed-effects regression with 97% power was identified and differences in pre-post scores were detected at 5% significance level. Adjusting for family random effects, breast cancer literacy scores increased for all participants recruited (P-value = 0.0004) suggesting that the process was feasible. Analysis of focus groups and action plans indicated that participants valued the instruction and planned to act upon it. This experience with kin keepers and their families offers encouragement that the theoretical model and its community-based delivery can continue to enhance scholarship dedicated to ameliorating health care disparities.